Singularity Pulse  
     
  Finance

Decimal Point Analytics Predicts Financial Status of Households Using Images

Decimal Point Analytics, a financial research and analytics company, has implemented Deep Learning Network technology for Image Analysis. It has launched beta version of “ImagePulse”, a niche product which makes predictions of financial status of households by processing their images. It uses its multi-layer convolutional neural networks for making predictions. Shailesh Dhuri, Executive Director, is confident and expects it to grow over time and provide valuable insights to sectors like banking, microfinance, government departments, insurance etc.

 
  Image  
 

Commerzbank To Use AI Intelligence to Automate Analyst Reports

Germany's second largest bank, Commerzbank, is doing research on AI with an aim of automating the process of writing analyst reports. The bank has partnered with a content-automation company called Retresco to develop the final algorithm. The current technology is far from providing ready reports for clients, but the company is confident as the equity research reports which review quarterly earnings are structured in similar manner. As present, the algorithm can produce 75% of what human analysts would write.

Finance  
  Image  
  Finance

AI Publishes Its Own Crypto Whitepaper After Reading 100 ICO Papers

Clay Space, founder of Blockspace Media, recently fed an AI system with 100 ICO (Initial Coin Offering) whitepapers and made AI produce one on its own. The AI system had no prior knowledge on cryptocurrency or ICOs. The resulting paper has striking similarities in terms of resembles and structure. The paper has listed uses of blockchain, process of purchasing ICOs, including normal public sale processes. Other sections of paper talked about use of funds and on team members and advisors.

 
  Image  
 

AI Algorithm Able to Create Periodic Table of Elements on Its Own

Stanford physicists have developed a new AI algorithm called “Atom2Vec” which organized the periodic table of elements and created its own periodic table in a matter of few hours. The program learnt to distinguish between different atoms after analyzing a list of chemical compound names from an online database. The unsupervised AI algorithm used natural language processing to cluster elements based on their chemical properties. The algorithm is modelled on AI program Google created to parse natural language.

Research  
  Image  
  Health Care

AI Powered Virtual Personal Trainer to Help with Perfect Squat

Kaia Health, a digital therapy company, has developed an app which will serve as a personal trainer for users. The app uses motion-tracking technology and AI to help users perform a perfect squat. Users are guided by a virtual assistant called Kaia. It tracks 16 key points on body via phone’s camera stream. It then contrasts the observed post with the ‘ideal’ predefined pose. It then compares matrices including relative positions of limbs and joints, and the angle between them. The app is currently available for free and available on Apple store.

 
  Image  
 

AI Algorithm to Aid Self-driving Cars to Traverse Across Intersection in Minimum Time

SRM Institute of Science and Technology in India has done a research on new technology tagged as 'Vehicle to Vehicle' (V2V) and ‘Vehicle to everything’ communication. This allows moving cars and objects to communicate with each other. Research was aimed at finding algorithm which can allow cars “to traverse across the intersection in minimum time whilst also ensuring that none of the vehicles collide with other vehicles and/or obstacles". It also raises concern over data privacy as it talks about the need of route sharing amongst cars for optimal solutions.

Miscellaneous  
  Image  
  Sports

IBM’s Watson To Generate Automated Wimbledon Highlights

IBM is going to leverage its super computer Watson’s AI algorithms to produce automated highlights of matches during Wimbledon for its fans. All England Lawn Tennis Club (AELTC) has partnered with IBM to work on deploying AI solutions to improve its business efficiency. Currently, Watson has been made to learn recognizing players’ emotional gestures and sound levels of crowd cheers to gauge potential highlight worthy point being played at any of its six main courts. In 2017, AI-generated video highlights garnered more than 14 million total views.

 
  Image  
 

“Team” Of Five AI Algorithms to Challenge Humans in Online Gaming

AI research company OpenAI has designed a group of five neural networks called “OpenAI Five” which works as a team while playing real-time computer strategy game Dota2. In its current version, it can beat a team of five human amateur players. The algorithm was first trained as individual in one-vs-one matches and later as team by playing against itself and past versions of itself. In a single day of its training, it gained experience worth 180 years. None of the five algorithms were allowed to communicate with each other.

Online Gaming  
  Image  
 
Image
“Quantum Computing, a revolution in technology, has shown great promise towards solving complex computing problems currently outside the capabilities of current computers. Though in its infancy, we at Decimal Point Analytics strongly believe that it is going to grow exponentially in the near future. It holds the potential not only to boost the AI revolution but also transform the way data is synthesized. The articles on Quantum computing will give an insight into the recent developments in this space.”
 
  Image  
  Quantum Computing

University of Melbourne Claims 60 Qubit Quantum Computing Simulation

Quantum computing uses qubits represented by atoms to encode and process data in multiple states to solve a problem. University of Melbourne simulated the output of a 60-qubit quantum computer. The simulation, otherwise, would have required up to 18000 petabytes of classical computer memory. Pawsey supercomputing facility in Perth conducted the simulation. Currently, the most advanced quantum computing computer is capable of simulating a random state of 50 qubits only.

 
  Image  
 

Alteration of Electrons to Improve Quantum Computing Efficiency

Purdue University has come up with two new ways of manipulating electronic interactions. This will help in improving the design and scaling up the capacity of quantum computing devices. the first method discovers that hydrostatic pressure can tune the interaction between paired and lined-up electron arrangements, providing desired properties and additional design options. The second method focuses on interaction of an electron's spin with its motion on surface of silicon, analyzing potential use of various materials to help stabilize the quantum realm.

Quantum Computing  
  Image  
 
 
 
     
  Banner  
 
Contact us:

5A, B-Wing, Trade Star Building, J. B. Nagar, Andheri-Kurla Road, Andheri (East), Mumbai - 400 059, Maharashtra, India
W: www.decimalpointanalytics.com  
E: info@decimalpointanalytics.com   T: +91 22300 15200